ca-leginfo-mcp
MCP server for California statutes and legislation, enabling verification of current statutory text, tracking pending bills that would amend or repeal sections, and retrieving legislative history from 1993 onward using official legislative bulk data.
README
ca-leginfo-mcp
An MCP server for California statutes and legislation, built on the Legislature's official public bulk-data channel (downloads.leginfo.legislature.ca.gov).
Lets an attorney's AI tool:
- Verify statutory text from source — current text, history note, and effective date for any section of the 29 California codes or the Constitution
- Flag pending bills that would amend, repeal, or add a given section
- Retrieve legislative history — from a code section to its enacting and amending bills and their committee/floor analyses, back to 1993
Status
All four build phases complete (ingest pipeline, archives, MCP server,
deployment). The server runs on Railway with artifacts in Cloudflare R2,
rebuilt nightly (deploy/):
https://server-production-602d.up.railway.app/mcp (streamable HTTP)
ingest/builds both corpus artifacts from the pubinfo bulk zips:current.db(~40 s nightly: current law + session, FTS, parsed bill→section refs, analysis text, sanity gate) andarchive.db(4.7 GB, 18 sessions 1989–2023: 283k bill version texts, 371k text-extracted analyses, votes, vetoes, ~1M parsed section refs, per-session coverage matrix).server/serves both over MCP (mcp2.0): seven tools — section text/search, pending-bill flagging, bill lookup, analyses with full text, legislative history (initiative and constitutional-amendment aware), chapter→bill pivot. Every response carries extract dates and a source note; era gaps come back as explicit coverage statements. Run locally:ca-leginfo-server --current-db … --archive-db …(stdio;--transport httpfor streamable HTTP).deploy/is the Railway + R2 layer:boot.pysyncs the artifacts from R2 and serves (/healthreports freshness); a nightly cron (nightly.py, 10:00 UTC) rebuilds current.db from the freshest pubinfo daily zip, gates it against last night's artifact, and uploads only on PASS. Simple per-client rate limiting on HTTP.
The feasibility spike that preceded the build is preserved in
spike/ — see SPIKE_FINDINGS.md.
Data source
All data derives from the California Legislature's public "pubinfo" bulk
downloads — the sanctioned distribution channel for this data (the
capublic schema published at downloads.leginfo.legislature.ca.gov).
No scraping is involved.
This is an unofficial convenience mirror. Statute text refreshes weekly (bills daily); every response carries its extract dates. For court filings, verify against the official publication.
Disclaimer
This tool retrieves law; it does not give legal advice.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。